本文根據兩波電話訪問資料，探討問題型式差異對民眾政治知識的影響。除檢視受訪者在不同題目型式的答題情形外，也同時探討問題型式對不同主題的政治知識的影響，最後並探討影響受訪者政治知識表現的因素。本文研究發現：不同題目型式顯著影響受訪者答題表現。相同文字語句測量以及控制其他可能影響因素下，無論是何種面向的政治知識，受訪者在封閉式題型的表現都比開放式題型佳，題型設計效果在政治人物面向的區辨力最大，在制度面向的差距最小。其次，如同預期，教育程度、媒體接觸、年齡等變項顯著影響個人政治知識高低，除了制度面向知識外，問卷型式對政治知識程度的影響力更甚於教育程度。第三，比較兩波資料發現，藉由電視和網路關注選舉新聞的民眾，在開放題型的表現較佳，顯示愈多的新聞關注有助提升選民的政治知識。最後，本文進一步探討問卷型式和性別間交互關係，性別在政治知識差距的大小和問卷型式有顯著關係，採選擇題型式可能使性別間的政治知識差距縮小。

By analyzing data from two cross-sectional telephone interviews, this article aims to examine the effect of question format on measurement of political knowledge in Taiwan. It first tackles the response differences among open-ended and closed-ended questions by grouping in terms of question topic. It then further studies the factors that play a role in influencing an individual’s political knowledge. The findings suggest that question format significantly affects individuals’ responses. Answering the same question wording but in different question formats, the respondents obtained higher scores on closed-ended than open-ended questions. Knowledge questions on the topic of politicians had a higher difference, while the knowledge questions on policies dis¬played a narrower gap. Secondly and unsurprisingly, variables such as the level of education, media exposure, and age played important roles in one’s level of political knowledge. Thirdly, those who paid more attention to election news through the media of television and internet performed better with open-ended questions. This finding illustrates that media exposure positively facilitated political knowledge. Lastly, a further study was done by examining the interaction effects between question format and gender. Different question for¬ mats highlight the gender gap. Closed-ended questions narrowed the knowledge gap between males and females, whereas, to some extent, a larger knowledge gap was found in open-ended questions. Question format indeed yields response effects.